How to use Safety Capacitors - What are they? 

Thank you - and welcome to the team!

Engineers do design with capacitors. I think what you meant is "design with safety capacitors".

Thanks so much!

You bet! Thanks for the feedback!

Thanks so much!

You are welcome! I'm glad this was informative. Just one of those things that doesn't come up often.

You Bet! I'm glad it helped;)

My pleasure! I appreciate you!

That's great! Let me know if there is something specific that you would like to see.

I really appreciate that! Very nice of you! I had a lot of fun at HNS;) Great to have you here!

That’s great - thanks!

Thanks Dan! I appreciate you!

Thank you! I really appreciate you!

Thanks! Glad it was helpful!

Thanks so much!

Glad it was helpful! I appreciate you!

Glad you enjoyed it

Thanks so much! Great to have you on board!

I'm so glad! I appreciate you!

That's awesome! I really enjoyed it there!

That's great! Thanks for that feedback:)

Thank you - I need it;)

Thanks BeamFinder! Yes, I will do an EMI filter design tutorial. It is interesting how little is done in some cases, but some filtering is better than none I suppose;) Oh my - I've forgotten about those old caps! I think the capacitor could be the most explosive thing on the board;) I'm always afraid of a cap going nuclear! I'm glad you were not hurt!

When I was an audiophile - and trying to build the best audio quality - we loved Roederstein capacitors and resistors. I use these poly caps when making speaker crossovers or anything in the signal path. I'm sure that they would have given you a cool sample kit. If I find the sample books - I'll show them in a video. Really cool the way they show the specs.

Thanks for the great question! The value would depend on the noise that you want to get rid of. If you have both caps try the biggest one;)

@@KissAnalog thank you, you're best :-)

Great point - I'm so glad that you made it! Yes, these are great caps - just big - but high quality. The film caps would be great for speaker crossovers or the generator that you mentioned. Polypropylene caps are a great choice for many applications.

There is typically one on each line to chassis. The largest value for this is 4700 pF. It is advisable to replace both. Poly caps would be a great alternative.

Bert van Lelieveld Thanks Bert! Glad to hear from you! It is very interesting to hear how things are done in other countries.

That’s a great question. I think it is because it is possible to have an undetected short or fault to ground. If this happens then the X type capacitor is not as safe as a Y type capacitor. Hope that makes sense.

@@KissAnalog thanks for your reply. But if there was a short to ground/earth then the RCD at the consumer unit would trip. I'm in the UK so probably using different terminology to North America. Do you think the restriction is to cater for old wiring installations which may not have been upgraded for many years and therefore not have an RCD fitted? Or for a faulty RCD (one where the contacts have stuck)? i think I may have answered my own question lol.

Great Question! No the damping filter (Cx in series with an R) can keep an input from ringing in some cases - depending on input impedance. A dampened input is not a bad idea. Depending on the power level - the filter can be quite simple.

You bet Robert - great to hear from you!

Thank you! The smaller caps are probably ok. If it noisy and interferes with other devices then you will have to get larger caps:)

Thanks David! I appreciate you!

Thank you! Much appreciated!

Thank you - I appreciate you! Is the X cap blown up, or also the Y caps? Was it shattered or burnt looking? I wonder if it was slowly damaged over time or burst from a lightning or surge? Is it a ceramic or poly (disc or box cap, or possibly Orange drop (molded type))? Is it X1, probably is. You can replace it with any type of X1 with a larger value as well. This filters noise from the blender so not to put the motor noise on the ac lines. They would likely use the smallest possible to save the cost. The Same goes for the Y caps, use Y1 and the same value any type or smaller value. The Y caps add leakage current on the ground and you don’t want too much of that. Don’t add TVS unless they are rated 10,000 Amps, but if they are rated in Watts they are not strong enough and will blow up eventually. The resistor value is too small, or was it in series with the X cap? If right across the lines it is probably k ohms. Did you measure it? Is it physically large?

@@KissAnalog Thank you for your fast reply, It is a cylinder combo XY capacitor in one component with five wires sticking out, the middle one (differing color, like transparent insulation hose) is supposed to be connected to the motor's stator iron it was an aluminium foil type with wax paper insulation layers rolled in between, this type of capacitors are known to explode and to pop their contents with force, which is my case, since the wax vaporizes and expands much more intensely and quickly than any plastic foil material will. I think it damaged over time because of deterioration of the insulation, this type of capacitor is so rare and I didn't find any with the same values, i just found one with same shape but with different value it was 0.1uF + 2x2700pF but i didn't purchase it because the 0.1 is far and not the same as 0.2uF and i need to connect it in parallel with other 0.1uF to obtain the 0.2uF. so i got from that store a 0.22uF X2 MKP type and i intend to connect it only without any Y capacitor. sorry about the resistors it is 1.65 Mega ohm and it is ordinary physical resistor

I don’t think I’ve seen a capacitor like that. The poly caps used to be expensive, but they seem very reasonable now. Is the X2 cap rated for the voltage? The MKP are self healing too, so probably a nice upgrade. Thanks for sharing this;)

@@KissAnalog yes, 250 Volt I will send a photo for you

Thanks Bjorn! Yes, some safety caps are actually marked Y1 X1, or Y2 X1 for instance. So, I've always thought it was OK to replace an X type with a Y type but not the other way around. I'll have to talk to a capacitor vendor to get the latest. I've worked in a lot of labs that use what they call cheater plugs to get rid of the chassis connection. The power supply in the scope is isolated, so the BNC connector will tie the scope ground to the signal ground, but it cuts the connection from the chassis to the Neutral of the power input. I could see it adding noise though. For surge protection, a capacitor can absorb some low energy spikes, but for more protection it would be better to use large TVS diodes or MOVs - if done correctly;) I have used large capacitors with small resistors, like say 22 uF with a 5 ohm resistor to absorb some transients. I like this better than a TVS as it is more robust. But if we are talking lightning protection - then that is not enough. I do like the failure mode of a TVS more than a MOV. The TVS will short then maybe crack open - where the MOV can start a fire. So, if you are looking to protect from a transient coming into your diff probes from the circuit that they are monitoring - then they are high impedance, so a TVS should work great. The TVS also has a parasitic capacitance, so if this is a problem - you can place a signal diode with low capacitance in series with the TVS diode. I will be doing some more videos on surge protection, so stay tuned;)

@@KissAnalog thank you for the quick reply. Regarding the scope ground, the powersupply is isolated, but in the modern scopes where the chassis is made of plastic, the incoming groundwire is connected to the negative part of the BNC connectors. If you then use a "cheater plug" or a isolation transformer, the incoming mains will leak for 50% to the negative part of the BNC connectors and has influence on the quality of measurements. The EMI filter is useally as follow : Neutral wire - X cap - Ground - X cap - Hot wire , but if the ground is not connected to the mains outlet (but still connected inside the scope to the negative part of the BNC connectors, then the neutral and hot wire will leak this small current to the BNC connectors. After reading on the internet about removing the ground pin for eliminating damage to the scope, i though this was the solution, because i have a modern Atten 60Mhz scope, with plastic chassis and no danger for getting a hot chassis. After some testing with this, i came to the conclusion that there was arround 120V on the negative part of the BNC connectors (i didn't feel anything when by touching it, but a tester with neon bulb did light up). And the measurements with the scope gave false readings. Then i did experiments with different small value X1 capacitors in series between the groundpin of the mains inlet (inside the scope), to the ground of the scope, to enduce this leaking to ground (has with connected earthpin was originally), but also the possibility for limiting high currents when accidentally the ground of the probe is connected to a hot wire. That worked, but even with 15a30V remaining on the negative part of the BNC connectors, it will give not accurate measurements. I even placed a white powerled in series, this powerled glows for a minutes and then failed due the high voltage. The same above about removing the earth to a scope occurs when a scope is placed behind a isolation transformer. That's why it is not recommended to place a scope behind a isolation transformer, but only the DUT (Device Under Test). Also when a DUT is placed behind a isolation transformer, and if safety caps are used inside the DUT, this leaking of the safety caps will then also give false readings on the scope if the ground of the DUT is not 0Volts.

I was asked by a major vendor of set top boxes once to tell them how they should run power; 3 prong or 2 prong power cord, . They had several manufacturers making boxes for them and they wanted to have them all follow one method. After thinking about it I wrote a white paper explaining why I thought it should be a 2 wire cord. The housing was plastic so no need to use ground and some people still don’t have ground and some people use extension cords that remove the ground. But the BNC will have a floating charge on it due to the Y caps, and UL requires transient protection to ground if 3 conductor. Just another potential that is not good if ground is not used. So I agree with you. I should experiment with my scope;)

I believe that they both go open. How could they rate it as X1Y1, if X would short and Y would open?

@@kgsalvage6306 The X capacitor is placed across line and neutral, in case of a surge or transient, or failure, they go short and pop a fuse or circuitbreaker, where the Y capacitor between line and ground goes open, when in case of failure there is no risk for shock (hot chassis). www.tecategroup.com/products/capacitors/ac-safety-capacitors.php

Thanks so much!!

You Bet! It is a Gamin Fenix ;)

I do plan to do a video on amplifier types:)

Wow, thanks!

Thanks for this great feedback! Were those clear looking capacitors kind of a gold or silver under the clear plastic? We need to use X1 and Y1 caps on the 240Vac applications. The X2 and Y2 are for the 120 VAC lines.

Just had a HP 8640B billow fire and smoke after running 30 mins on 230V. Y cap had failed spectacularly.PME271Y cap

Thanks Geir! If you have a ground wire in the power cable, then you could try Y caps, but if not - then it could actually be a safety issue. I'll have to cover this in a video:)

You're welcome! And thank you for your feedback;)

​@@KissAnalog I had been looking for information on X2/Y2 caps. I have been in electronics since the mid-70's but never studied them. All of the industry instruments I worked (still work) on are portable or non-portable have power entry modules that in addition to fuses have the line filters built-in. Now I plan to play with restoring old tube radios and other tube equipment and needed to learn about the capacitors tied to the power line. Your video was perfect and what I was looking for. The only other thing I was taugh from elsewhere: Y1/X1 = Industrial, Y2/X2 = consumer. You cover a little bit about that. One thing that possibly could be added is the failure mode of each type. I think Y caps open, X Caps short. I would be interested in a demonstration video of Y/X cap failures and how they are built to open or short. Maybe not important but interesting.

Thanks Bert! Are you talking about the GFCI circuits?

Hi Jun-Hwa that is a great question! You should be able to test it with a capacitor measurement meter - like the one on your multimeter. If there is no physical damage and the meter reads OK in capacitance - then it probably is OK. What is the reason you suspect the capacitor? You can safely replace an X1 with two X2 capacitors. Try to get double the value so in series it is back to the original value. Let me know how it goes...

@@KissAnalog Thank you for your reply and apologized for not backing to you sooner. I did a simple heat air blow onto the PSU and it turned on. Saw somewhere on youtube and tested it myself. So I started my project by getting all replacement caps (without cap measurement meter, a proper soldering gun and solder sucker pump :P). I replaced all caps, burned circuit board and RIP to old PSU lol. In the process, I found out that PSU manufacture used higher grade solder alloy probably mixed silver (melting point is 420C). Removed old caps took much long including X1. PSU didn't turn on T-T. Long story short, my friend gave me his spare 1000W PSU and PC is up running :) Few days later, I ordered new digital clap meter with cap measurement meter later on and tested all old caps from PSU. Caps (Chemi-con Japanese caps) turned out values are in acceptable range after 12 years of use. X1 cap also fine. Could have waited for the new meter before starting the task but it was a good learning curve.

Thanks for the great feedback! The idea about the motor start caps is a great idea! Here's a link to an audio playlist: ru-vid.com/group/PLZ0_iMoMBSsmeLp-eZfrBKMGOxrNQT_Mo

@@KissAnalog Ha! I see you've done a Gainclone too! This definitely takes me a while back! Re. motor start caps: I've been trying to find out about their general applicability/corresponding de-rating without success for ages. Not that I'd recommend it, but very often these can be had for nothing when scavenging parts from old equipment. At the same time, sometimes people throw away equipment just because of bad motor cap (which of course they wouldn't know about). All the same, these caps are cheap, and maybe sometimes free, when their regular equivalents (?) cost 10x more.

I agree. We need to do a good video to better explain this type of capacitor.

LOL, thank you Stephen!

Hi Glenn - it has been awhile! Good to hear from you. Yes, the SMT is making it hard for me to solder - well I mean mostly impossible;)

Thanks for your feedback! I do like the TDK capacitors - especially the ceramics;)

Thanks for the great question Denny! Leakage current should come out to just under the limit of 0.5 mA, so it looks like you might have got one of the numbers to the wrong power.

@@KissAnalog - Wow. You are right! Pico is -12, not -9... Duh!!

LOL

Thanks for the great question! This is very intersting - as it is reactive power - in a sense. However, it is actually a load so it will dissipate a very small amount of power. The X2 caps are well worth it - as it is an infantly small amount of dissipation. Surge protectors typically have MOV protection devices which are also capacitive, so they will draw a bit of current - but not as much as the X caps. I hope this helps.

@@KissAnalog I have another SPD that is only MOV based and has no X2 capacitor. The current is below my digital ammeter's measurement capabilities.

@@KissAnalog can you point me to any technical information that gives numbers for active vs reactive power of X2 capacitors (at 240 VAC 50Hz)? I did some googling and came up empty handed. Perhaps it is possible to derive this from the datasheets, but if so then I don't yet have the knowledge to do so.

1 is needed for higher voltage. Leakage is a function of the capacitor value.

Thanks for the great question. I'm sorry it took me so long to get to it. I'm having trouble keeping up so I try to get subs first then work my way through the list;) I'll do a video on this.

@@KissAnalogThank you very much my respectable sir....

OK that's interesting. This device is the same - it is looking for current in the line and neutral to be nearly the same. In the states the new standard requires a GFCI (Ground Fault Circuit Interruptor) to be used in any location that there can be moisture. It was just for swimming pools and such, then it became necessary for bathrooms and kitchens, but now I think it is for any unfinished spaces or 'for any space that may have moisture'. And then there is a Class A which is used for residential and has a limit of 5 mAmps and a class B which is meant to be more safety for the equipment and has a limit of 30 mAmps. Sometimes it is referred to as the GFI without the C.

In the USA we have to have a GFI on the plug of all hair driers because of safety due to "wet" locations like a bathroom where most people dry their hair after a shower. Eddie explains what a GFCI and GFI does above as a reply!

LOL - but so true;)

Capacitors never used nF, but as the Art of Electronics is slipping away, so to are the conventions. In this case it wasn't really a big deal maybe. The uF was used to standardize the units, it made for several advantages, but maybe today the use of nF makes sense...I'm not sure...

Yes, I think the idea is that they fail open. However, I don’t think I’ve ever seen one fail;)

Great question Don! I didn't know that they did, can you give me a part number and I'll look it up? Thanks!

@@KissAnalog cpc.farnell.com/evox-rifa/pzb300mc23r30/capacitor-delta-network-0-15uf/dp/CA08124?mckv=sx5FCDr5d_dc|pcrid|490691434319|kword||match||plid||slid||product|CA08124|pgrid|48077370878|ptaid|pla-369572310414|&CMP=KNC-GUK-CPC-SHOPPING&s_kwcid=AL!5616!3!490691434319!!!network}!369572310414!&gclid=CjwKCAjwv_iEBhASEiwARoemvJ53EX_COiDB4gGTr0WKik04XBjd90MMmZXDecUiunRtCfb4mVh39xoCDq8QAvD_BwE

Sounds like a safe control. Thanks for sharing that - great info!

Thanks for asking! Do you mean electrically or visually? They will show a marking X2, X1, Y1, or Y2 on them.

Yep - this is time consuming;) I hear the trails calling...

Thanks for your feedback! Did you have the inverter chassis grounded, or tied to your 12V board (was it just a board or was it in a metal box)? Have you seen this grounding Rant? ;) ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-sgp-0xR8Voc.html

I agree, but it is used so often I used the 'safety' in this title. They should be safe for line voltage because of their construction. But, the names 'RFI' and 'XY' are also used. It is unsafe to use normal poly caps on line voltages, so they made these with extra insulation between the capacitive metal elements to be 'safe'.

LOL what do you mean?

Sorry you didn't learn anything. You can hit the fast forward... What do you think we gain by stalling? It takes time to do a video - so if viewers click off - it doesn't help us.

@@KissAnalogsome people just like to complain 😅